Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Dominance Statistics: A Simulation Study on the d Statistic

Dominance Statistics: A Simulation Study on the d Statistic Cliff (1993) has proposed the use of a measure of effect size alternative to traditionalmean differences: δ {? = Pr(xi1 > xj2) - Pr(xi1 < xj2)}which, taken a pair of values, xi1 and xj2, from the first and second populations respectively, gives the probability that the value from the first populationis higher than the one from the second, minus the probability of the inverse, which is,in fact, an alternative measure of effect size. In this paper we test, using computersimulation techniques, the robustness and power of Wilcoxon-Mann-Whitney's U,and Cliff's d statistics with samples of equal and unequal sizes under homoscedasticityand heteroscedasticity. We conclude that d test looks like a good alternative to traditionalmean tests, not only because it is not so much restricted by the parametric assumptions, but because it tests an effect size indicator which is nearer to researchers' real interests. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Quality & Quantity Springer Journals

Dominance Statistics: A Simulation Study on the d Statistic

Loading next page...
1
 
/lp/springer-journals/dominance-statistics-a-simulation-study-on-the-d-statistic-g0BHoQ00kM

References (15)

Publisher
Springer Journals
Copyright
Copyright © 2003 by Kluwer Academic Publishers
Subject
Social Sciences; Methodology of the Social Sciences; Social Sciences, general
ISSN
0033-5177
eISSN
1573-7845
DOI
10.1023/A:1024469826841
Publisher site
See Article on Publisher Site

Abstract

Cliff (1993) has proposed the use of a measure of effect size alternative to traditionalmean differences: δ {? = Pr(xi1 > xj2) - Pr(xi1 < xj2)}which, taken a pair of values, xi1 and xj2, from the first and second populations respectively, gives the probability that the value from the first populationis higher than the one from the second, minus the probability of the inverse, which is,in fact, an alternative measure of effect size. In this paper we test, using computersimulation techniques, the robustness and power of Wilcoxon-Mann-Whitney's U,and Cliff's d statistics with samples of equal and unequal sizes under homoscedasticityand heteroscedasticity. We conclude that d test looks like a good alternative to traditionalmean tests, not only because it is not so much restricted by the parametric assumptions, but because it tests an effect size indicator which is nearer to researchers' real interests.

Journal

Quality & QuantitySpringer Journals

Published: Oct 17, 2004

There are no references for this article.